The exploration and interaction of humanoid robots with theenvironment through tactile sensing is an important task forachieving truly autonomous agents. Recently much researchhas been focused on the development of new technologiesfor tactile sensors and new methods for tactile exploration.Edge detection is one of the tasks required in robots andhumanoids to explore and recognise objects. In this work wepropose a method for edge and plane classification with abiomimetic iCub fingertip using a probabilistic approach. TheiCub fingertip mounted on anxy-table robot is able to tap andcollect the data from the surface and edge of a plastic wall.Using a maximum likelihood classifier thexy-table knowswhen the iCub fingertip has reached the edge of the object. Thestudy presented here is also biologically inspired by the tactileexploration performed in animals.
|Title of host publication||Proceedings of the Post-Graduate Conference on Robotics and Development of Cognition , 2012|
|Number of pages||4|
|Publication status||Published - 20 May 2012|
|Event|| Post-Graduate Conference on Robotics and Development of Cognition - Lausanne, Switzerland|
Duration: 10 Sep 2012 → 12 Sep 2012
|Conference||Post-Graduate Conference on Robotics and Development of Cognition|
|Period||10/09/12 → 12/09/12|